This is a competing renewal for the Johns Hopkins/Rio de Janeiro ICIDR, Innovative Approaches to Control TB. For the past 5 years, we have developed the infrastructure and human resources to conduct outstanding clinical research on TB control. We have performed two trials of novel TB control strategies, and a series of lab studies assessing new diagnostic techniques. We now propose to build on this track record to address three critically important research questions in global TB control: optimal case finding strategies, treatment shortening, and prevention of MDR TB. Project 1 is a community-randomized trial comparing a strategy of DOTS to an augmented strategy of DOTS and active case finding (ACF) through outreach workers in the community. We will randomize 20 discrete districts in the northern areas of Rio de Janeiro, matched by baseline TB incidence, to DOTS vs. DOTS with annual intensive ACF. We will measure reduction of TB incidence in these communities as the primary endpoint, and hypothesize that DOTS/ACF will reduce TB incidence 40% more than DOTS alone after 4 years. Project 2 is a randomized clinical trial of a moxifloxacin (Moxi) as a treatment shortening agent for newly diagnosed TB. We will randomize 526 patients with pulmonary TB to receive Moxi/rifampin/PZA/ethambutol for 4 months vs. a control regimen of INH/RIF/PZA/EMB for 6 months. We hypothesize that the Moxi regimen will be better tolerated, have higher completion rates, and will be non-inferior in curing disease. Project 3 is a randomized trial of two treatment regimens for latent TB infection of contacts of patients with MDR TB. We will randomize 220 patients to receive moxi alone for 6 months vs. ofloxacin/EMB for 6 months, with outcomes of safety, tolerablility and effectiveness. We hypothesize that moxi alone will have few adverse events, higher completions rates, and similar efficacy vs. ofloxacin/EMB. Administrative and biostats/data management cores will support these studies. These trials will contribute to improving our understanding of the control of TB worldwide. PROJECT 1: A Clustered Randomized Trial of DOTS Plus Active Case Finding (Chaisson, R.) PROJECT 1 DESCRIPTION (provided by applicant): In most countries, tuberculosis (TB) case finding is passive, relying on symptomatic patients to seek health care on their own initiative. The primary objective of active case finding (ACF) is to identify all active cases of TB within a geographic location. Early detection of active TB can reduce morbidity, mortality, and transmission in a population. The goal of this project is to determine whether TB active case finding (ACF) in conjunction with a Directly Observed Therapy, Short course (DOTS) results in lower incidence of TB compared to routine DOTS alone. This community randomized trial will be conducted in 20 Rio de Janeiro communities within two planning areas (AP) with high rates of TB where community health workers and health units are undergoing intensive training to implement DOTS programs. This will be a collaborative effort between the Municipal Health Secretariat of Rio de Janeiro and the Johns Hopkins University Center for Tuberculosis Research. We will conduct a clustered randomized trial to compare the impact of two case detection and treatment strategies on the community incidence of TB. The 20 communities within 2 APs will be randomized to routine DOTS based on passive case detection or DOTS plus door-to-door active case finding (DOTS-ACF), where TB cases are actively detected by community outreach. These 20 communities will utilize community health workers who routinely collect health information in these communities. We hypothesize that the average TB incidence in Years 3, 4 and 5 will be 40% less in the DOTS-ACF arm compared to the DOTS arm and that treatment completion rates will not differ between the two arms of the study.